CN111080525B - Distributed image and graphic primitive splicing method based on SIFT features - Google Patents
Distributed image and graphic primitive splicing method based on SIFT features Download PDFInfo
- Publication number
- CN111080525B CN111080525B CN201911319821.0A CN201911319821A CN111080525B CN 111080525 B CN111080525 B CN 111080525B CN 201911319821 A CN201911319821 A CN 201911319821A CN 111080525 B CN111080525 B CN 111080525B
- Authority
- CN
- China
- Prior art keywords
- image
- computing node
- primitive
- library
- master
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 239000013598 vector Substances 0.000 claims abstract description 12
- 230000004927 fusion Effects 0.000 claims abstract description 8
- 239000000284 extract Substances 0.000 claims abstract description 6
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 239000000463 material Substances 0.000 claims description 18
- 239000000523 sample Substances 0.000 claims description 17
- 238000007667 floating Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 230000002194 synthesizing effect Effects 0.000 claims description 2
- 238000000605 extraction Methods 0.000 abstract description 5
- 238000005457 optimization Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- HPTJABJPZMULFH-UHFFFAOYSA-N 12-[(Cyclohexylcarbamoyl)amino]dodecanoic acid Chemical compound OC(=O)CCCCCCCCCCCNC(=O)NC1CCCCC1 HPTJABJPZMULFH-UHFFFAOYSA-N 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
A distributed image and primitive splicing method based on SIFT features divides a main computing node and sub computing nodes by establishing a distributed heterogeneous parallel system; image preprocessing is carried out by utilizing the acquired image samples to be spliced, and the preprocessed image samples are uploaded to a main computing node to serve as a global image library; initializing a local image library and a primitive cell library of each sub-computing node; each sub-computing node pre-extracts characteristic points of all images in a local image library by using a SIFT algorithm, stores the calculated high-dimensional characteristic vectors into corresponding data structures, and marks the images in the local image library as a state to be spliced; each sub-computing node is locally spliced and circulated; global primitive fusion update broadcast; and controlling the task flow of the main computing node. The splicing method optimizes and improves the splicing computation complexity aiming at the aspects of feature extraction, matching and the like of mass images, and performs algorithm design optimization and module function realization by applying parallelism in a scene operation flow.
Description
Technical Field
The invention belongs to the technical field of computer image processing, and particularly relates to a distributed image and primitive splicing method based on SIFT features aiming at a massive image splicing application scene.
Background
The image stitching technology is a hot spot research object in the field of computer image processing, and the most intuitive image description is a technology for stitching a group of images with overlapping areas into a panoramic image with higher resolution and wider visual angle, and the technology begins from the data processing requirements in the fields of satellite remote sensing, map databases and the like. With rapid progress in information technology, computer and microelectronic technology, the requirements of more industry fields such as medical research, robot research and development, military reconnaissance aerial photography and the like on image splicing technology are higher and higher, and especially, the popularization of equipment such as smart phones, digital cameras, unmanned aerial vehicles, VR virtual reality and the like is higher and higher, so that the technical threshold and cost for obtaining high-definition images are lower and lower. However, because the shooting device is limited by the imaging optical principle, resolution and other performance parameters of the device, the visual field range of a single picture is still limited, so that people expect to splice mass image materials obtained by shooting for multiple times to obtain similar panoramic images.
At present, an image stitching method based on feature matching is widely adopted, and a Scale-invariant feature transform (Scale-invaiant feature transform) SIFT feature matching method proposed by David Lowe has revolutionary influence on the image stitching industry, and various improvements and variant algorithms such as SURF, ORB, BRISK based on the principle of the method appear later, so that the method has advantages and disadvantages. The SIFT feature matching method has strong robustness to translation, rotation, scaling and affine transformation of the image, but the principle is complex, the extraction of feature points, the generation of feature descriptors and the calculation load of the Euclidean distance of high-dimensional vectors in matching are large, the timeliness of the SIFT feature matching method is limited, and the inefficiency defect is more obvious especially in an application scene aiming at massive image stitching.
Aiming at the mass image splicing task based on the SIFT method, a distributed parallel mechanism is naturally provided, and the performance improvement method is adopted in a large amount in the unmanned aerial vehicle aerial photography field. However, these methods often have certain limitations in implementation, for example, before parallelization processing of image stitching, the stitching sequence of the images needs to be known in advance. This additional "sequential" prior information is more demanding in terms of pre-processing of the acquired image material and also limits the scalability of the related parallel solution.
Disclosure of Invention
The invention aims to solve the technical problem of a distributed image and primitive splicing method based on SIFT features, and the splicing method aims to introduce a GPU heterogeneous parallel mechanism to further accelerate feature extraction, feature vector generation and high-dimensional feature vector Euclidean distance calculation in traditional SIFT feature matching.
In order to solve the technical problems, the invention is realized by the following steps:
a distributed image and primitive splicing method based on SIFT features specifically comprises the following steps:
1) Computing environment and data preparation
1-1) establishing a usable distributed heterogeneous parallel software and hardware system, and definitely dividing a master computing node and a slave computing node;
1-2) acquiring mass image sample materials to be spliced;
1-3) image preprocessing, namely image denoising and gray level equalization conversion, so that the illumination brightness of the obtained image sample materials is consistent and the image characteristics are clear and obvious;
1-4) uploading all the preprocessed image sample materials to a master computing node master and taking the master computing node master as a global image library;
2) Distributed heterogeneous parallel algorithm module initialization
2-1) initializing a local image library of each sub-computing node: the main computing node mater distributes image sample materials in the global image library to each sub computing node noder uniformly to serve as a local image library, and different sub computing nodes noder are distributed to task data sets belonging to the main computing node mater and ensure load balancing;
2-2) initializing a primitive cell library: the master computing node master randomly selects one image sample material set from the uploaded image sample material set, an initial primitive library is constructed, the image is split in a primitive cell reference system according to a preset primitive scale constant, coordinate numbers (xIndex, yIndex) are generated for each split primitive, the coordinate numbers are stored in a global primitive library of the master computing node master, and then the master computing node master broadcasts the initial primitive cell library to the child computing nodes noder to serve as a local primitive library;
2-3) each sub-computing node pre-extracts characteristic points of all images in a local image library by using a SIFT algorithm, stores the calculated high-dimensional characteristic vector information into a corresponding data structure, and marks all images in the local image library as a state to be spliced;
3) Distributed heterogeneous parallel algorithm load computation
3-1) sub-computing node noder local splicing circulation;
3-2) fusion updating and broadcasting of global graphics primitives;
3-3) master computing node master task flow control.
Compared with the prior art, the invention has the beneficial effects that:
the first, the distributed algorithm in this patent is directed against the massive image splicing scene, its required image comes from the shooting material of the visual field scope, do not require the splice of order among the image, namely there is not the restriction requirement such as arranging the naming to the image material in order, even do not require the image that must be the same time batch shooting, the main computing node master that sets up in the distributed splicing algorithm will distribute all image material to each sub-computing node noder at random, equally, and the image material used for setting up the initial primitive storehouse is also selected at random.
In the local splicing circulation on each sub-computing node noder, an automatic polling mechanism is adopted, namely, splicing attempt is carried out on all unmarked images to be spliced in a local image library and a global and local dynamically-expanded primitive library, manual intervention operation is not needed in the whole processing flow, and the characteristic matching calculation between the images is controlled by an algorithm.
Thirdly, aiming at the computation-intensive operations such as relevant feature description extraction, feature point pair matching and the like in the SIFT method, each sub-computing node noder can configure a GPU graphics accelerator, and further acceleration processing is carried out by utilizing a CUDA heterogeneous parallel programming method, so that the distributed algorithm solution has an easily-extensible attribute with dynamic tuning performance.
Drawings
FIG. 1 is a schematic diagram of the initialization flow of the distributed heterogeneous parallel algorithm module of the present invention.
Fig. 2 is a schematic diagram of a local splicing loop of a sub-computing node noder according to the present invention.
FIG. 3 is a schematic diagram of a global primitive fusion update and broadcast process according to the present invention.
Detailed Description
The following describes the embodiments of the present invention in further detail with reference to the drawings and specific examples.
A distributed image and primitive splicing method based on SIFT features specifically comprises the following steps:
1) Computing environment and data preparation
1-1) establishing a usable distributed heterogeneous parallel software and hardware system, and definitely dividing a master computing node and a slave computing node;
1-2) acquiring mass image sample materials to be spliced;
1-3) image preprocessing, namely image denoising and gray level equalization conversion, so that the illumination brightness of the obtained image sample materials is consistent and the image characteristics are clear and obvious;
1-4) uploading all the preprocessed image sample materials to a master computing node master and taking the master computing node master as a global image library;
2) The distributed heterogeneous parallel algorithm module is initialized, as shown in figure 1,
2-1) initializing a local image library of each sub-computing node: the main computing node mater distributes image sample materials in the global image library to each sub computing node noder uniformly to serve as a local image library, and different sub computing nodes noder are distributed to task data sets belonging to the main computing node mater and ensure load balancing;
2-2) initializing a primitive cell library: the master computing node master randomly selects one image sample material set from the uploaded image sample material set, an initial primitive library is constructed, the image is split in a primitive cell reference system according to a preset primitive scale constant, coordinate numbers (xIndex, yIndex) are generated for each split primitive, the coordinate numbers are stored in a global primitive library of the master computing node master, and then the master computing node master broadcasts the initial primitive cell library to the child computing nodes noder to serve as a local primitive library;
2-3) each sub-computing node pre-extracts characteristic points of all images in a local image library by using a SIFT algorithm, stores the calculated high-dimensional characteristic vector information into a corresponding data structure, and marks all images in the local image library as a state to be spliced;
3) Distributed heterogeneous parallel algorithm load computation
3-1) sub-computing node noder local splicing circulation; as shown in fig. 2
3-1-1) each sub-computing node updates local primitive cell library data, extracts characteristic point information of all primitives by using a SIFT algorithm and computes high-dimensional characteristic vectors;
3-1-2) selecting an image in a state to be spliced from a local image library, and performing characteristic point matching attempt on the image and all the primitives in the primitive cell library;
3-1-3) if the Euclidean distance matched with SIFT feature points is not satisfied and is smaller than the splicing matching requirement of the set threshold, turning to the next image to be spliced to perform the same treatment;
3-1-4) finding out the best matching graphic element from the result meeting the matching requirement, wherein the best matching graphic element is obtained by calculating the Euclidean distance of high-dimensional feature vectors of matching pairs of the image and the graphic element, sorting from low to high, counting feature point pairs smaller than a certain distance threshold value in each matching pair, and selecting the situation with the maximum number of the point pairs as the best matching;
screening and filtering characteristic point pairs, wherein the specific steps are that firstly, a maximum Euclidean distance threshold value maxThreshold (80.0) and a point pair upper limit number threshold value maxMatchNumThreshold (500) are set, and meanwhile, a floating coefficient index is set, and the initial value of the floating coefficient index is 0.01; taking the product of the floating coefficient index and the maximum distance value max_distance in the matching pair as a temporary distance threshold value tempThreshold, wherein the temporary distance threshold value does not exceed a maximum Euclidean distance threshold value, calculating the number of point pairs with the matching pair Euclidean distance smaller than the maximum Euclidean distance threshold value, and increasing the floating coefficient index value by 0.01 when the number of point pairs is smaller than the upper limit number threshold value of the point pairs, and repeating the process for 100 times; if the temporary distance threshold exceeds the maximum Euclidean distance threshold or the number of the point pairs smaller than the temporary distance threshold exceeds the upper limit number of the point pairs, stopping iteration, taking the temporary distance threshold at the moment as the Euclidean distance threshold of final screening and filtering, and eliminating all the feature matching point pairs higher than the Euclidean distance threshold.
3-1-5) carrying out further fine screening on the characteristic matching point pairs screened in the step 3-1-4) by utilizing a RANSAC random sampling consistency algorithm, so as to calculate and obtain an accurate coordinate transformation matrix of the projection mapping of the selected image to the optimal matching primitive space;
3-1-6) completing image and primitive splicing according to the coordinate transformation matrix, splitting the result in a primitive cell coordinate system based on the position of the optimal primitive to generate a corresponding primitive coordinate number [ xIndex, yIndex ], and returning to the step 3-1-1).
If all the new primitive numbers split in the step 3-1-6) are the same as the original primitive numbers, fusion processing is needed, and the redundancy of local primitive cell library data and subsequent repeated calculation are avoided. This patent marks the effective pixel region in the image through using the mask mode, and the primitive fusion rule is: and (3) two graphic element cells with the same number, wherein the areas marked with 0 at the same pixel positions of the corresponding mask masks represent invalid pixels (the values are set to 0), the areas marked with 255 by the masks represent overlapping, the fusion processing is carried out by using a linear average fusion method, and the effective pixels marked with 255 by the mask are directly used under other conditions to form a final fusion graphic element, and meanwhile, the mask of the graphic element after fusion is updated and calculated according to the effectiveness of the final pixel.
3-2) fusion updating and broadcasting of global graphics primitives; as shown in fig. 3
3-2-1) each sub-computing node iterates through a node loop until no new primitive is generated in the primitive cell library on the node (equivalent to no new splicing success event occurs in the local area);
3-2-2) if the number of images in the state to be spliced in the node local image library is reduced, which indicates that the node contributes to the update of the global primitive cell library, sending the node local primitive cell library back to the master computing node master, and then waiting for the update message sent by the master computing node master;
3-2-3) if the number of images in the state to be spliced in the node local image library is kept unchanged, which indicates that the node does not contribute to the updating of the global primitive cell library, a special null event message is sent back, unnecessary node communication is reduced, and the updating message transmitted by the master of the main computing node is waited for.
3-3) Master task flow control for Master computing node
3-3-1) after the master of the main computing node completes the distribution of the global image library and the broadcasting operation of the global primitive cell library, waiting for and receiving the information transmitted by each sub computing node, and counting the splicing state of the distributed images and the primitives on each sub computing node noder;
3-3-2) if the child computing node noder sends back a non-null event message, fusing the node local primitive cell library sent back into the global primitive cell library, and broadcasting the final fused result again to enable each child computing node noder to enter the next round of splicing iteration;
3-3-3) if all the sub-computing nodes noder send back a null event message, indicating that the splicing task of the distributed image and the graphic element is finished, broadcasting a splicing end message by the master node master, and synthesizing all the graphic elements in the global graphic element cell library into a large graph according to the coordinates of the corresponding graphic element cell coordinate system, and outputting the large graph as a final result;
3-3-4) after receiving the splicing end broadcast message, each sub-computing node noder releases the relevant resources applied by the node and exits.
The steps of establishing a scale space in the related SIFT feature point extraction operation, including Gaussian pyramid, dog pyramid, scale space extremum detection, extremum point accurate positioning and the like, are realized by using a mature OpenCV open source library.
Aiming at the Euclidean distance of the high-dimensional feature vector calculated in the step 3-1-4), a GPU acceleration mechanism is adopted, and a specific implementation scheme is CUDA programming development specifications issued by NVIDIA company.
The foregoing is merely illustrative of the embodiments of this invention and it will be appreciated by those skilled in the art that variations may be made without departing from the principles of the invention, and such modifications are intended to be within the scope of the invention as defined in the claims.
Claims (5)
1. A distributed image and primitive splicing method based on SIFT features is characterized in that: the method comprises the following steps:
1) Computing environment and data preparation
1-1) establishing a usable distributed heterogeneous parallel software and hardware system, and definitely dividing a master computing node and a slave computing node;
1-2) acquiring mass image sample materials to be spliced;
1-3) image preprocessing, namely image denoising and gray level equalization conversion, so that the illumination brightness of the obtained image sample materials is consistent and the image characteristics are clear and obvious;
1-4) uploading all the preprocessed image sample materials to a master computing node master and taking the master computing node master as a global image library;
2) Distributed heterogeneous parallel algorithm module initialization
2-1) initializing a local image library of each sub-computing node: the main computing node mater distributes image sample materials in the global image library to each sub computing node noder uniformly to serve as a local image library, and different sub computing nodes noder are distributed to task data sets belonging to the main computing node mater and ensure load balancing;
2-2) initializing a primitive cell library: the master computing node master randomly selects one image from the uploaded image sample material set, an initial primitive library is constructed, the image is split in a primitive cell reference system according to a preset primitive scale constant, coordinate numbers (xIndex, yIndex) are generated for each split primitive, the coordinate numbers are stored in a global primitive library of the master computing node master, and then the master computing node master broadcasts the initial primitive cell library to the child computing nodes noder to serve as a local primitive library;
2-3) each sub-computing node pre-extracts characteristic points of all images in a local image library by using a SIFT algorithm, stores the calculated high-dimensional characteristic vector information into a corresponding data structure, and marks all images in the local image library as a state to be spliced;
3) Distributed heterogeneous parallel algorithm load computation
3-1) sub-computing node noder local splicing circulation;
3-2) fusion updating and broadcasting of global graphics primitives;
3-3) master computing node master task flow control.
2. The distributed image and primitive stitching method based on SIFT features of claim 1, wherein: the specific flow steps of the step 3-1) are as follows:
3-1-1) each sub-computing node updates local primitive cell library data, extracts characteristic point information of all primitives by using a SIFT algorithm and computes high-dimensional characteristic vectors;
3-1-2) selecting an image in a state to be spliced from a local image library, and performing characteristic point matching attempt on the image and all the primitives in the primitive cell library;
3-1-3) if the Euclidean distance matched with SIFT feature points is not satisfied and is smaller than the splicing matching requirement of the set threshold, turning to the next image to be spliced to perform the same treatment;
3-1-4) finding out the best matching graphic element from the result meeting the matching requirement, wherein the best matching graphic element is obtained by calculating the Euclidean distance of high-dimensional feature vectors of matching pairs of the image and the graphic element, sorting from low to high, counting feature point pairs smaller than a certain distance threshold value in each matching pair, and selecting the situation with the maximum number of the point pairs as the best matching;
3-1-5) carrying out further fine screening on the characteristic matching point pairs screened in the step 3-1-4) by utilizing a RANSAC random sampling consistency algorithm, so as to calculate and obtain an accurate coordinate transformation matrix of the projection mapping of the selected image to the optimal matching primitive space;
3-1-6) completing image and primitive splicing according to the coordinate transformation matrix, splitting the result in a primitive cell coordinate system based on the position of the optimal primitive to generate a corresponding primitive coordinate number [ xIndex, yIndex ], and returning to the step 3-1-1).
3. The distributed image and primitive stitching method based on SIFT features of claim 1, wherein: the specific flow steps of the step 3-2) are as follows:
3-2-1) each sub-computing node iterates in a noder loop until no new primitive is generated in the primitive cell library on the node;
3-2-2) if the number of images in the state to be spliced in the node local image library is reduced, which indicates that the node contributes to the update of the global primitive cell library, sending the node local primitive cell library back to the master computing node master, and then waiting for the update message sent by the master computing node master;
3-2-3) if the number of images in the state to be spliced in the node local image library is kept unchanged, which indicates that the node does not contribute to the updating of the global primitive cell library, a special null event message is sent back, unnecessary node communication is reduced, and the updating message transmitted by the master of the main computing node is waited for.
4. The distributed image and primitive stitching method based on SIFT features of claim 1, wherein: the specific flow steps of the step 3-3) are as follows:
3-3-1) after the master of the main computing node completes the distribution of the global image library and the broadcasting operation of the global primitive cell library, waiting for and receiving the information transmitted by each sub computing node, and counting the splicing state of the distributed images and the primitives on each sub computing node noder;
3-3-2) if the child computing node noder sends back a non-null event message, fusing the node local primitive cell library sent back into the global primitive cell library, and broadcasting the final fused result again to enable each child computing node noder to enter the next round of splicing iteration;
3-3-3) if all the sub-computing nodes noder send back a null event message, indicating that the splicing task of the distributed image and the graphic element is finished, broadcasting a splicing end message by the master node master, and synthesizing all the graphic elements in the global graphic element cell library into a large graph according to the coordinates of the corresponding graphic element cell coordinate system, and outputting the large graph as a final result;
3-3-4) after receiving the splicing end broadcast message, each sub-computing node noder releases the relevant resources applied by the node and exits.
5. The distributed image and primitive stitching method based on SIFT features of claim 2, wherein: the specific flow steps of the step 3-1-4) are as follows:
firstly, setting a maximum Euclidean distance threshold value maxThreshold and a point-to-upper limit quantity threshold value maxMatchNumTohreshold, and simultaneously setting a floating coefficient index, wherein the initial value of the floating coefficient index is 0.01; taking the product of the floating coefficient index and the maximum distance value max_distance in the matching pair as a temporary distance threshold value tempThreshold, wherein the temporary distance threshold value does not exceed a maximum Euclidean distance threshold value, calculating the number of point pairs with the matching pair Euclidean distance smaller than the maximum Euclidean distance threshold value, and increasing the floating coefficient index value by 0.01 when the number of point pairs is smaller than the upper limit number threshold value of the point pairs, and repeating the process for 100 times; if the temporary distance threshold exceeds the maximum Euclidean distance threshold or the number of the point pairs smaller than the temporary distance threshold exceeds the upper limit number of the point pairs, stopping iteration, taking the temporary distance threshold at the moment as the Euclidean distance threshold of final screening and filtering, and eliminating all the feature matching point pairs higher than the Euclidean distance threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911319821.0A CN111080525B (en) | 2019-12-19 | 2019-12-19 | Distributed image and graphic primitive splicing method based on SIFT features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911319821.0A CN111080525B (en) | 2019-12-19 | 2019-12-19 | Distributed image and graphic primitive splicing method based on SIFT features |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111080525A CN111080525A (en) | 2020-04-28 |
| CN111080525B true CN111080525B (en) | 2023-04-28 |
Family
ID=70315881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911319821.0A Active CN111080525B (en) | 2019-12-19 | 2019-12-19 | Distributed image and graphic primitive splicing method based on SIFT features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111080525B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111401482B (en) * | 2020-04-29 | 2024-03-19 | Oppo广东移动通信有限公司 | Feature point matching method and device, equipment and storage medium |
| CN114769021B (en) * | 2022-04-24 | 2022-11-25 | 广东天太机器人有限公司 | Robot spraying system and method based on full-angle template recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105590086A (en) * | 2014-11-17 | 2016-05-18 | 西安三茗科技有限责任公司 | Article antitheft detection method based on visual tag identification |
| CN109493278A (en) * | 2018-10-24 | 2019-03-19 | 北京工业大学 | A kind of large scene image mosaic system based on SIFT feature |
| CN109636714A (en) * | 2018-08-30 | 2019-04-16 | 沈阳聚声医疗系统有限公司 | A kind of image split-joint method of ultrasonic wide-scene imaging |
| CN110232673A (en) * | 2019-05-30 | 2019-09-13 | 电子科技大学 | A kind of quick steady image split-joint method based on medical micro-imaging |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100986809B1 (en) * | 2008-07-17 | 2010-10-08 | 인하대학교 산학협력단 | The Method of Automatic Geometric Correction for Multi-resolution Satellite Images using Scale Invariant Feature Transform |
| JP5470607B2 (en) * | 2010-09-29 | 2014-04-16 | 株式会社日立製作所 | Computer system and method for aligning images / charts |
| US20120301014A1 (en) * | 2011-05-27 | 2012-11-29 | Microsoft Corporation | Learning to rank local interest points |
| CN102496033B (en) * | 2011-12-05 | 2014-06-18 | 西安电子科技大学 | Image SIFT feature matching method based on MR computation framework |
| CN102968777B (en) * | 2012-11-20 | 2015-01-14 | 河海大学 | Image stitching method based on overlapping region scale-invariant feather transform (SIFT) feature points |
| CN103426153B (en) * | 2013-07-24 | 2016-01-20 | 广州地理研究所 | The quick joining method of a kind of unmanned aerial vehicle remote sensing images |
| CN105245841B (en) * | 2015-10-08 | 2018-10-09 | 北京工业大学 | A kind of panoramic video monitoring system based on CUDA |
| US10249047B2 (en) * | 2016-09-13 | 2019-04-02 | Intelligent Fusion Technology, Inc. | System and method for detecting and tracking multiple moving targets based on wide-area motion imagery |
| CN106599917A (en) * | 2016-12-09 | 2017-04-26 | 西北大学 | Similar image duplicate detection method based on sparse representation |
| CN107067370A (en) * | 2017-04-12 | 2017-08-18 | 长沙全度影像科技有限公司 | A kind of image split-joint method based on distortion of the mesh |
| CN107665479A (en) * | 2017-09-05 | 2018-02-06 | 平安科技（深圳）有限公司 | A kind of feature extracting method, panorama mosaic method and its device, equipment and computer-readable recording medium |
| CN109801212A (en) * | 2018-12-26 | 2019-05-24 | 南京信息职业技术学院 | A kind of fish eye images joining method based on SIFT feature |
| CN109951922A (en) * | 2019-04-03 | 2019-06-28 | 深圳市南方海擎科技有限公司 | A kind of light source light spectrum self-correcting and self-test closed-loop control system and its control method |
| CN110225315A (en) * | 2019-07-12 | 2019-09-10 | 北京派克盛宏电子科技有限公司 | Electric system screen monitored picture fusion method |
-
2019
- 2019-12-19 CN CN201911319821.0A patent/CN111080525B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105590086A (en) * | 2014-11-17 | 2016-05-18 | 西安三茗科技有限责任公司 | Article antitheft detection method based on visual tag identification |
| CN109636714A (en) * | 2018-08-30 | 2019-04-16 | 沈阳聚声医疗系统有限公司 | A kind of image split-joint method of ultrasonic wide-scene imaging |
| CN109493278A (en) * | 2018-10-24 | 2019-03-19 | 北京工业大学 | A kind of large scene image mosaic system based on SIFT feature |
| CN110232673A (en) * | 2019-05-30 | 2019-09-13 | 电子科技大学 | A kind of quick steady image split-joint method based on medical micro-imaging |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111080525A (en) | 2020-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Surrounding vehicle detection using an FPGA panoramic camera and deep CNNs | |
| CN110853075B (en) | Visual tracking positioning method based on dense point cloud and synthetic view | |
| CN107679537B (en) | A kind of texture-free spatial target posture algorithm for estimating based on profile point ORB characteristic matching | |
| CN109509255B (en) | Tagged map construction and space map updating method and device | |
| CN111968129A (en) | Instant positioning and map construction system and method with semantic perception | |
| CN104778656B (en) | Fisheye image correcting method based on spherical perspective projection | |
| CN111028151B (en) | Point cloud data splicing method based on graph residual neural network fusion | |
| CN111080525B (en) | Distributed image and graphic primitive splicing method based on SIFT features | |
| CN107945217B (en) | Image characteristic point pair rapid screening method and system suitable for automatic assembly | |
| CN112288628B (en) | Aerial image splicing acceleration method and system based on optical flow tracking and frame extraction mapping | |
| Cheng et al. | Omnidirectional depth extension networks | |
| CN111553939A (en) | Image registration algorithm of multi-view camera | |
| WO2021035627A1 (en) | Depth map acquisition method and device, and computer storage medium | |
| CN112734914A (en) | Image stereo reconstruction method and device for augmented reality vision | |
| CN113379748A (en) | Point cloud panorama segmentation method and device | |
| Gao et al. | Multi-source data-based 3D digital preservation of largescale ancient chinese architecture: A case report | |
| KR101998318B1 (en) | 360-degree Image Generation Apparatus and Method for Synchronizing Time and Position of Object | |
| CN117237623B (en) | Semantic segmentation method and system for remote sensing image of unmanned aerial vehicle | |
| CN111626307A (en) | Plant image real-time splicing method based on L-ORB algorithm | |
| CN115713678A (en) | Arrow picture data augmentation method and system, electronic device and storage medium | |
| CN115937452A (en) | Map reconstruction method and system based on distributed computation | |
| Cui et al. | 3D reconstruction with spherical cameras | |
| Kawasaki et al. | Super‐resolution omnidirectional camera images using spatio‐temporal analysis | |
| Zhang et al. | Image mosaic of bionic compound eye imaging system based on image overlap rate prior | |
| CN116740488B (en) | Training method and device for feature extraction model for visual positioning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |